alamb commented on code in PR #10766:
URL: https://github.com/apache/datafusion/pull/10766#discussion_r1623492772


##########
datafusion/core/tests/parquet/arrow_statistics.rs:
##########
@@ -592,58 +739,180 @@ async fn test_timestamp_diff_rg_sizes() {
     }
     .run();
 
+    Test {
+        reader: reader.build().await,
+        // mins are [2020-01-01T12:01:01+11:00, 2020-01-01T12:11:01+11:00, 
2020-01-11T12:02:01+11:00]
+        expected_min: Arc::new(
+            TimestampNanosecondArray::from(vec![
+                TimestampNanosecondType::parse("2020-01-01T01:01:01"),
+                TimestampNanosecondType::parse("2020-01-01T01:11:01"),
+                TimestampNanosecondType::parse("2020-01-11T01:02:01"),
+            ])
+            .with_timezone(tz),
+        ),
+        // maxes are [2020-01-02T01:01:01+11:00, 2020-01-11T01:01:01+11:00, 
2020-01-12T01:01:01+11:00]
+        expected_max: Arc::new(
+            TimestampNanosecondArray::from(vec![
+                TimestampNanosecondType::parse("2020-01-02T01:01:01"),
+                TimestampNanosecondType::parse("2020-01-11T01:01:01"),
+                TimestampNanosecondType::parse("2020-01-12T01:01:01"),
+            ])
+            .with_timezone(tz),
+        ),
+        // nulls are [1, 2, 1]
+        expected_null_counts: UInt64Array::from(vec![1, 2, 1]),
+        // row counts are [8, 8, 4]
+        expected_row_counts: UInt64Array::from(vec![8, 8, 4]),
+        column_name: "nanos_timezoned",
+    }
+    .run();
+
     // micros
     Test {
         reader: reader.build().await,
-        expected_min: Arc::new(Int64Array::from(vec![
-            1577840461000000,
-            1577841061000000,
-            1578704521000000,
+        // mins are [2020-01-01T01:01:01, 2020-01-01T01:11:01, 
2020-01-11T01:02:01]

Review Comment:
   these comments seem a little redundant given that the actual timestamp 
string appears in the test as well



##########
datafusion/core/tests/parquet/mod.rs:
##########
@@ -403,9 +437,13 @@ fn make_timestamp_batch(offset: Duration) -> RecordBatch {
         schema,
         vec![
             Arc::new(arr_nanos),
-            Arc::new(arr_micros),
-            Arc::new(arr_millis),
-            Arc::new(arr_seconds),
+            Arc::new(arr_nanos_timezoned),
+            Arc::new(arr_micros.clone()),

Review Comment:
   I wonder if these really need to be cloned 🤔 
   
   ```suggestion
               Arc::new(arr_micros),
   ```



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to